Highlights from the ATLAS Experiment

Available online at www.sciencedirect.com

Nuclear Physics A 982 (2019) 8–14 www.elsevier.com/locate/nuclphysa

XXVIIth International Conference on Ultrarelativistic Nucleus-Nucleus Collisions (Quark Matter 2018)

Highlights from the ATLAS experiment

Iwona Grabowska-Bold on behalf of the ATLAS Collaborationa,

aAGH University of Science and Technology, Krak´ow, Poland

Abstract This report provides an overview of the new results obtained by the ATLAS Collaboration at the LHC, which were presented at the Quark Matter 2018 conference. These measurements were covered in 12 parallel talks, one flash talk and 11 posters. In this document, a discussion of results is grouped into four areas: electromagnetic interactions, jet quenching, quarkonia and heavy-flavour production, and collectivity in small and larger systems. Measurements from the xenon-xenon collisions based on a short run collected in October 2017 are reported for the first time.

Keywords: ATLAS experiment, quark-gluon plasma, photon-induced processes, jet quenching, quarkonia production, heavy-ﬂavour production, collectivity in small systems, xenon-xenon collisions

1. Introduction

In addition to proton-proton (pp) physics, the ATLAS Collaboration [1] participates in the heavy- ion (HI) programme which has been carried out at the LHC since 2010. Lead-lead (Pb+Pb) and proton- lead (p+Pb) collisions were provided at the centre-of-mass energies of 2.76 TeV, 5.02 TeV for Pb+Pb, and 5.02 TeV for p+Pb. In October 2017 a short period with xenon-xenon (Xe+Xe) collisions was taken. This opened up an opportunity of studying impact of different geometries on a broad range of observables. More- over, the HI programme is supplemented with measurements in the pp system, which serve as a reference to disentangle initial- from final-state effects.

2. Electromagnetic interactions in the QGP

The strong electromagnetic ﬁeld associated with highly boosted nuclei at the LHC can be utilised to study the scattering of the quasi-real photons emitted coherently from the nuclei as they pass by next to each other. These are so-called Ultra-Peripheral Collisions (UPC). In the previous measurements of exlusive production of di-muon pairs [2] and light-by-light scattering [3], ATLAS already demonstrated that photon

https://doi.org/10.1016/j.nuclphysa.2018.08.024 0375-9474/© 2018 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). I. Grabowska-Bold / Nuclear Physics A 982 (2019) 8–14 9

fluxes emerging from Pb beams are well modelled by the STARlight generator. Also in those measurements, Δφ an acoplanarity distribution (α = 1 − π , where Δφ is a distance in azimuth between final-state particles) was measured and proved to be powerful in discriminating between signal and background processes. Very recenty ATLAS has explored a potential of observing events from exclusive production of γγ → μ+μ− contributing to a sample of inelastic minimum-bias Pb+Pb collisions [4]. After evaluating and re- moving the contribution from background sources, the azimuthal angle (Δφ) and transverse momentum (pT) correlations between the muons are measured as a function of collision centrality. Figure 1 shows the background-subtracted acoplanarity distribution in different centrality intervals. Each distribution is normalised to unity over its measured range. The > 80% distribution with a significant contribution from UPC events is plotted in each panel for comparison. A clear, centrality-dependent broadening is seen in the acoplanarity distributions when compared to the > 80% interval. The corresponding distribution from the γγ → μ+μ− MC samples is also shown. The MC α distributions show almost no centrality dependence, indicating that the broadening evident in the data is notably larger than that expected from detector effects. One potential source of modification is the final-state interaction of the produced leptons with the electric charges in the QGP. Assuming that the broadening of the α distribution results from transfers of a small amount of pT to each muon, in the 0–10% centrality interval that scale, assumed to be the RMS momentum transfer to each final-state muon in the transverse plane, is evaluated to amount to 70 ± 10 MeV.

Fig. 1. Background-subtracted acoplanarity distribution (α) in four centrality intervals in Pb+Pb data [4]. A comparison to the STARlight calculation for γγ → μ+μ− is also shown. The distributions are normalised to unity over their measured range.

3. Jet quenching

Using the large statistics of the 2015 Pb+Pb data, ATLAS finalised measurements of inclusive jet nuclear modification factor (RAA) [5], as well as jet fragmentation functions [6]. Those results provide very detailed studies of inclusive jet production as a function of jet pT, rapidity (y) and centrality in comparison to the reference pp data collected at 5.02 TeV. The RAA evaluated as a function of jet pT for two centrality intervals 0–10% and 30–40% is presented in the left panel of Fig. 2. The RAA value is obtained for jets with |y| < 2.1 and with pT between 80–1000 GeV. A clear suppression of jet production in central Pb+Pb collisions relative to pp collisions is observed. In the 0-10% centrality interval, RAA is approximately 0.45 at pT = 100 GeV, and is observed to grow slowly (quenching decreases) with increasing jet pT, reaching a value of 0.6 for jets with pT around√ 800 GeV. In the same figure, the RAA values at 5.02 TeV are compared with the previous measurements at sNN = 2.76 TeV. The two measurements agree within their uncertainties in the overlapping pT region. The apparent reduction of the size of systematic uncertainties in the new measurement is possible thans to large samples of pp and Pb+Pb data collected during the same LHC running period. Further insight in jet quenching can be obtained by studying jet fragmentation functions. The right panel of Fig. 2 presents a ratio of transverse jet fragmentation functions D(pT)inPb+Pb to those extracted in pp collisions as a function of jet fragment pT for three jet pT intervals. The RD(pT) is above unity (enhancement) for low pT jet fragments, drops below unity for intermediate jet pT fragments (suppression) and becomes ff larger than unity again for jet fragment pT around 50 GeV. There is no significant di erence between RD(pT) for three jet pT intervals. A comparison between the data and the hybrid model calculation is also shown. 10 I. Grabowska-Bold / Nuclear Physics A 982 (2019) 8–14

2.5 jet AA ATLAS anti-k R = 0.4 jets |y| < 2.1 ATLAS | | < 2.1 anti- =0.4 jets t y kt R R ) 126 < p jet < 158 GeV 0 - 10%, s = 2.76 TeV [PRL 114 (2015) 072302] T NN T p 200 < p jet < 251 GeV 0 - 10%, s = 5.02 TeV ( NN 2 T 1 jet 30 - 40%, s = 2.76 TeV [PRL 114 (2015) 072302] D 316 < p < 398 GeV NN T Hybrid Model, R = 3 30 - 40%, sNN = 5.02 TeV R res T and luminosity uncer. 126 < p jet < 158 GeV 〈 AA〉 T 200 < p jet < 251 GeV 1.5 T 316 < p jet < 398 GeV T

0.5 Pb+Pb, = 5.02 TeV, 0.49 nb-1, 0-10% sNN 0.5 pp , s = 5.02 TeV, 25 pb-1 2 40 60 100 200 300 500 900 1 10 10 p [GeV] p [GeV] T T

Fig. 2. (Left) Inclusive jet RAA as a function of jet pT for jets with |y| < 2.1 in 0–10% and 30–40% centrality intervals compared to + the same quantity measured in 2.76 TeV Pb Pb collisions [5]. (Right) RD(pT) ratios for three jet pT ranges: 126–158 GeV (circles), 200–251 GeV (diamonds) and 316–398 GeV (crosses) compared with calculations from the hybrid model with Rres = 3 [6].

The model is able to describe the intermediate- and high-pT regions for jet fragments, while it fails in the low-pT region. The ATLAS Collaboration performed a preliminary measurement of the angular distribution of charged particles around the jet axis in 5.02 TeV Pb+Pb and pp data [7]. The measured yields are deﬁned as:

2 = 1 1 d nch(r), RD(pT) (1) Njet 2πr drdpT π where Njetis the total number of jets, 2 rdr is the area of the annulus at a given distance r from the jet axis (r = Δη2 +Δφ2 with Δη and Δφ being the relative differences between the charged particle and the jet axis, in pseudorapidity and azimuth respectively), dr is the width of the annulus and nch(r) is the number of charged particles within a given annulus. Results are presented as a function of Pb+Pb collision centrality, and both jet and charged-particle pT in the left panel of Fig. 3. Ratios of D(pT, r) distributions in Pb+Pb to those measured in pp collisions as a function of r for six charged-particle pT intervals spanning values between 1.6–63.1 GeV in 0–10% centrality, and for jet pT between 200–251 GeV are shown. The RD(pT,r) is above unity for all r values for charged particles with pT less than 4 GeV. For these particles, RD(pT,r) grows < . . < < . > . with increasing r for r 0 3 and is approximately constant for 0 3 r 0 6. For pT 4 0 GeV, RD(pT,r) is below unity and decreases with increasing r for r < 0.3 and is approximately constant for 0.3 < r < 0.6. The observed behaviour inside the jet (r < 0.4) agrees with the measurement of the inclusive jet fragmentation functions [6], where yields of the low-pT fragments are observed to be enhanced and yields of charged particles with intermediate pT are suppressed. The measured dependence of RD(pT,r) suggests that the energy lost by jets through the jet quenching process is being transferred to particles with pT < 4.0 GeV with larger radial distances. ATLAS also measures inclusive jet mass (m) divided by the jet transverse momentum [8]. This fully- unfolded measurement of the jet structure is sensitive to the angular and momentum correlations of the jet constituents. These correlations can be used to study modifications of jets in HI collisions, where they provide complementary information to previously measured jet fragmentation functions. The right panel of Fig. 3 presents RAA as a function of m/pT in 0–10% centrality for jet pT between 126–158 GeV. For all centrality bins, these values have no significant dependence on m/pT. They are also observed to be consistent with the inclusive jet RAA. A preliminary measurement of the balance between isolated photons and inclusive jets in pT in 5.02 TeV γ γ Pb+Pb and pp data is performed. Photons with pT > 63.1 GeV and |η | < 2.37 are paired inclusively with all jets that have pT > 31.6 GeV and |η| < 2.8 in the event. The transverse momentum balance given by the jet-to-photon pT ratio, xJγ, are measured for pairs with azimuthal opening angle |Δφ| > 7π/8. Distribu- tions of the per-photon jet yield (1/Nγ)(dN/dxJγ) are corrected for detector effects via a two-dimensional unfolding procedure and reported at the particle level for the first time. I. Grabowska-Bold / Nuclear Physics A 982 (2019) 8–14 11

Fig. 3. (Left) Ratios of D(pT, r) distributions in 0–10% Pb+Pb collisions to pp collisions as a function of angular distance r for jet pT of 200 to 251 GeV for six pT selections [9]. (Right) Jet RAA as a function of m/pT in 0–10% centrality for jet pT between 126–158 GeV [8].

+ γ = Figure 4 shows the measured xJγ distribution in five centrality intervals of Pb Pb collisions for pT 63.1 − 79.6 GeV in comparison to the xJγ distribution from pp collisions. The xJγ distributions in Pb+Pb collisions evolve smoothly with centrality. For peripheral collisions with centrality 50–80%, they are similar to those measured in pp collisions. However, in increasingly more central collisions, the distributions become systematically more modified. The xJγ distribution in the most central 0–10% events is so strongly modified that it is monotonically decreasing over the measured xJγ range and no peak is observed.

) 2 γ J 1.8 ATLAS Preliminary x 50-80% 30-50% -1 /d 1.6 pp 5.02 TeV, 25 pb N 1.4 Pb+Pb, 0.49 nb-1

)(d 1.2 γ 1 γ N p = 63.1-79.6 GeV 0.8 T

(1/ 0.6 pp (same each panel) 0.4 0.2 Pb+Pb )

γ 2 J

x 1.8 20-30% 10-20% 0-10%

/d 1.6

N 1.4 1.2 )(d γ 1 N 0.8

(1/ 0.6 0.4 0.2 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 x x x Jγ Jγ Jγ

Fig. 4. Photon-jet pT balance distributions (1/Nγ)(dN/dxJγ)inpp events (blue, reproduced on all panels) and Pb+Pb events (red) with each panel denoting a diﬀerent centrality selection [7].

-1 -1 -1 ATLAS Preliminary pp, 25 pb Xe+Xe, 3 μb Pb+Pb, 0.49 nb To probe physics of jet quenching in HI collisions with | |<2.5 s = 5.02 TeV s = 5.44 TeV s = 5.02 TeV AA η NN NN (extrapol. to 5.44 TeV) nuclei lighter than Pb, the transverse momentum asymme- R try of dijet pairs and production rates of charged particles 1 are measured by ATLAS with Xe+Xe collisions collected in October 2017 [10]. Figure 5 presents charged-hadron RAA as a function of pT for three centrality and Npart in- 0.3

+ Xe+Xe, N Pb+Pb, N tervals along with the measurement from the Pb Pb sys- 〈 part〉 〈 part〉 5-15%, 194 20-30%, 189 tem. The RAA is compared between Xe+Xe and Pb+Pb 30-40%, 84 40-50%, 87 ﬀ 55-70%, 24 60-80%, 23 data at 5.02 TeV. Even though they have di erent cen- 0.1 1 3310 0p [GeV] tralities, the Npart for the same pT intervals are compa- T rable. The Xe+Xe data shows more suppression than the + Pb Pb data in more central collisions, and less suppression Fig. 5. Charged-hadron RAA as a function of pT mea- in more peripheral collisions. sured in Xe+Xe collisions 5.44 TeV (closed mark- ers) and in Pb+Pb collisions at 5.02 TeV (open mark- ers) [10]. 12 I. Grabowska-Bold / Nuclear Physics A 982 (2019) 8–14

4. Quarkonia and heavy-ﬂavour production

The ATLAS Collaboration finalised a detailed study on prompt and non-prompt J/ψ and ψ(2S ) production and suppression at high pT in 5.02 TeV Pb+Pb and pp collisions [11]. The measurements of per-event yields, nuclear modification factors, and non-prompt fractions are performed in the di-muon decay chan- < μμ < . < < . nel for 9 pT 40 GeV in di-muon transverse momentum, and 2 0 yμμ 2 0 in rapidity. Strong suppression is found in Pb+Pb collisions for both prompt and non-prompt J/ψ, as well as for prompt and non-prompt ψ(2S ), increasing with event centrality. The suppression of prompt ψ(2S ) is observed to be stronger than that of J/ψ, while the suppression of non-prompt ψ(2S ) is equal to that of the non-prompt J/ψ within uncertainties, consistent with the expectation that both arise from b-quarks propagating through the medium. Despite prompt and non-prompt J/ψ arising from different mechanisms, the dependence of their nuclear modification factors on centrality is found to be similar. The left panel of Fig. 6 shows a pT-dependence of the nuclear modification factor for prompt J/ψ mesons reconstructed via the muon channel at 5.02 TeV in the 0–20% centrality bin. The RAA is at the level of 0.25 at pT = 9 GeV and tends to increase slowly with pT. The ATLAS measurement at high pT nicely complements the ALICE results for pT < 12 GeV that are also shown on the same figure. Recently the ATLAS Collaboration also evaluated elliptic flow of J/ψ with respect to the event plane in 5.02 TeV Pb+Pb collisions and presented preliminary results as a function of transverse momentum, rapidity and centrality [12]. It is observed that prompt and non-prompt J/ψ mesons have non-zero elliptic flow. Prompt J/ψ ν2 decreases as a function of pT, while non-prompt J/ψ ν2 is flat over the studied kinematical region. There is no dependence on rapidity or centrality observed. The right panel of Fig. 6 shows results for the ν2 as a function of pT for prompt and non-prompt J/ψ as measured by ATLAS compared with inclusive J/ψ at pT < 12 GeV, as measured by ALICE at 5.02 TeV, and prompt J/ψ at 6.5 < pT < 30 GeV, by CMS at 2.76 TeV. Despite different rapidity selections, the ATLAS data is found to be in reasonable agreement with the ALICE and CMS data in the overlapping pT region.

Fig. 6. (Left) Comparison of prompt J/ψ RAA measured in 5.02 TeV Pb+Pb collisions by ATLAS with the inclusive J/ψ RAA measured by ALICE [11]. (Right) ν2 as a function of pT for prompt and non-prompt J/ψ as measured by ATLAS compared with inclusive J/ψ at pT < 12 GeV, as measured by ALICE at 5.02 TeV, and prompt J/ψ at 6.5 < pT < 30 GeV by CMS at 2.76 TeV [12].

The ATLAS Collaboration also finalised a measurement of the production of muons from heavy-flavour decays in 2.76 TeV Pb+Pb and pp collisions [13]. Results are provided in the muon transverse momentum range 4 < pT < 14 GeV and for five centrality intervals. Backgrounds arising from in-flight pion and kaon decays, hadronic showers, and mis-reconstructed muons are statistically removed using a template-fitting procedure. The heavy-flavor muon differential cross-sections and per-event yields are measured in pp and Pb+Pb collisions, respectively. Figure 7 presents the heavy-flavour muon RAA as a function of pT. The RAA does not depend on pT within the uncertainties of the measurement. The RAA decreases between peripheral 40–60% collisions, where it is about 0.65, to more central collisions, reaching a value of about 0.35 in the 0–10% centrality interval. In Ref. [13] the azimuthal modulation of the heavy-flavor muon yields is also measured and the I. Grabowska-Bold / Nuclear Physics A 982 (2019) 8–14 13 associated Fourier coefficients νn for n=2, 3 and 4 are given as a function of pT and centrality. They vary slowly with pT and show a systematic variation with centrality which is characteristic of other anisotropy measurements, such as that observed for inclusive hadrons.

AA 1 ATLAS ATLAS R s = 2.76 TeV = 2.76 TeV NN sNN Pb+Pb, 0.14 nb -1 pp, 570 nb -1

0.5

0-10% Pb+Pb, 0.14 nb -1 -1 10-20% 20-30% pp, 570 nb | | < 1 30-40% η 40-60% |η| < 1 0 468101214 468101214 p [GeV] p [GeV] T T

Fig. 7. Heavy-ﬂavuor muon RAA as a function of pT for ﬁve centrality intervals in 2.76 TeV Pb+Pb collisions [13].

5. Collectivity in small and large systems One active area of ongoing research is investigation of the nature of the long-range ridge observed in two-particle correlations in small collision systems such as pp and p+Pb. To understand the multi-particle nature of the long-range collective phenomenon in those systems, the ATLAS Collaboration performed a measurement of symmetric cumulants scn,m{4} and asymmetric cumulants acn{3} which probe four- and three-particle correlations of two ﬂow harmonics νn and νm in 13 TeV pp, 5.02 TeV p+Pb, and 2.76 TeV peripheral Pb+Pb collisions [14]. The large non-ﬂow background from dijet production present in the standard cumulant method is suppressed using a method of subevent cumulants.

Fig. 8. The Nch dependence of sc2,3{4} (left), sc2,4{4} (middle) and ac2{3} (right) in 0.5 < pT < 5 GeV obtained for ppcollisions (solid circles), p+Pb collisions (open circles) and low-multiplicity Pb+Pb collisions (open squares) [14].

Figure 8 shows a comparison of cumulants for the three collision systems. The three panels present the results for sc2,3{4},sc2,4{4}, and ac2{3} for charged particles with 0.3 < pT < 3 GeV. These results indicate a negative correlation between ν2 and ν3 and a positive correlation between ν2 and ν4. Such correlation patterns have previously been observed in large collision systems, but are now confirmed also in the small collision systems, once non-flow effects are adequately removed in the measurements. In the multiplicity range covered by the pp collisions, Nch < 150, the results for symmetric cumulants sc2,3{4} and sc2,4{4} are comparable among the three systems. In the range Nch > 150, |sc2,3{4}| and sc2,4{4} are larger in Pb+Pb than in p+Pb collisions. The results for ac2{3} are similar among the three systems at Nch < 100, but they deviate from each other at higher Nch. The results for pp data are approximately constant or decrease slightly with Nch, while the p+Pb and Pb+Pb data shows significant increases as a function of Nch. The similarity between different collision systems and the weak dependence of these observables on the pT range and Nch, largely free from non-flow effects, provide an important input for understanding the space-time dynamics and the properties of the medium created in small collision systems. 14 I. Grabowska-Bold / Nuclear Physics A 982 (2019) 8–14

+ Using a data set of Xe Xe collisions col- ATLAS Preliminary 0.5

2. ATLAS also measured the modified Pearson’s correlation coefficient to quantify correlations between flow coefficients and mean pT of charged particles in the event using 5.02 TeV Pb+Pb data [16]. It can be used in further experimental studies to understand the underlying mechanism of QGP dynamics and constrain theoretical models attempting to describe them.

6. Summary

The ATLAS Collaboration presented many new results covering Pb+Pb, p+Pb, pp, and also data from the new Xe+Xe system collected for the ﬁrst time at the LHC. These measurements provide new information on electromagnetic interactions, the jet quenching, quarkonia and heavy-ﬂavour suppression, as well as comprehensive results which provide further insight into the collectivity phenomenon of small collision systems. This work was supported in part by Polish National Science Centre grant DEC-2016/23/B/ST2/01409, by the AGH UST statutory tasks No. 11.11.220.01/4 within subsidy of the Ministry of Science and Higher Education, and by PL-Grid Infrastructure. References

[1] ATLAS Collaboration, JINST 3 (2008) S08003. doi:10.1088/1748-0221/3/08/S08003. [2] ATLAS Collaboration, http://cdsweb.cern.ch/record/2157689, ATLAS-CONF-2016-025. [3] ATLAS Collaboration, arXiv:1702.01625 [hep-ex], Nature Physics 13 (2017) 852. [4] ATLAS Collaboration, arXiv:1806.08708 [nucl-ex], submitted to PRL. [5] ATLAS Collaboration, arXiv:1805.05635 [nucl-ex], submitted to PLB. [6] ATLAS Collaboration, arXiv:1805.05424 [nucl-ex], submitted to PRC. [7] ATLAS Collaboration, http://cdsweb.cern.ch/record/2318869, ATLAS-CONF-2018-010. [8] ATLAS Collaboration, http://cdsweb.cern.ch/record/2319867, ATLAS-CONF-2018-014. [9] ATLAS Collaboration, http://cdsweb.cern.ch/record/2318868, ATLAS-CONF-2018-009. [10] ATLAS Collaboration, http://cdsweb.cern.ch/record/2318588, ATLAS-CONF-2018-007. [11] ATLAS Collaboration, arXiv:1805.04077 [nucl-ex], submitted to EPJC. [12] ATLAS Collaboration, http://cdsweb.cern.ch/record/2319788, ATLAS-CONF-2018-013. [13] ATLAS Collaboration, arXiv:1805.05220 [nucl-ex], submitted to PRC. [14] ATLAS Collaboration, arXiv:1807.02012 [nucl-ex], submitted to PLB. [15] ATLAS Collaboration, http://cdsweb.cern.ch/record/2318870, ATLAS-CONF-2018-011. [16] ATLAS Collaboration, http://cdsweb.cern.ch/record/2318589, ATLAS-CONF-2018-008.